U.S. patent application number 13/371327 was filed with the patent office on 2013-08-15 for system and method for calculating power using contactless voltage waveform shape sensor.
The applicant listed for this patent is David ELROD, James E. OWEN, Daniel J. PARK. Invention is credited to David ELROD, James E. OWEN, Daniel J. PARK.
Application Number | 20130211751 13/371327 |
Document ID | / |
Family ID | 48946340 |
Filed Date | 2013-08-15 |
United States Patent
Application |
20130211751 |
Kind Code |
A1 |
PARK; Daniel J. ; et
al. |
August 15, 2013 |
System and Method for Calculating Power Using Contactless Voltage
Waveform Shape Sensor
Abstract
A system and method are provided for calculating power using a
voltage waveform shape measurement from a contactless sensor. An
electrically conductive medium carries alternating current (AC)
electrical current, associated with an AC voltage, from a source
node to a destination node. AC current is measured through the
electrically conductive medium. Using a contactless sensor, an AC
voltage waveform shape is measured. The power usage at the
destination node is calculated in response to the AC current
measurement, the measurement of the AC voltage waveform shape, and
an AC voltage potential. For simplicity, the AC current and AC
voltage waveform shape may both be measured at a first node located
between the source node and the destination node. The AC voltage
potential used in the power usage calculation may be an estimate,
an actual measurement, or a value supplied by an external source
(e.g., the power utility).
Inventors: |
PARK; Daniel J.; (Beaverton,
OR) ; OWEN; James E.; (Vancouver, WA) ; ELROD;
David; (Beaverton, OR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
PARK; Daniel J.
OWEN; James E.
ELROD; David |
Beaverton
Vancouver
Beaverton |
OR
WA
OR |
US
US
US |
|
|
Family ID: |
48946340 |
Appl. No.: |
13/371327 |
Filed: |
February 10, 2012 |
Current U.S.
Class: |
702/61 ;
324/76.11 |
Current CPC
Class: |
G01R 21/06 20130101 |
Class at
Publication: |
702/61 ;
324/76.11 |
International
Class: |
G01R 21/00 20060101
G01R021/00; G06F 19/00 20110101 G06F019/00 |
Claims
1. A method for calculating power using a voltage waveform shape
measurement from a contactless sensor, the method comprising:
providing an electrically conductive medium carrying alternating
current (AC) electrical current, associated with an AC voltage,
from a source node to a destination node; measuring the AC current
through the electrically conductive medium; using a contactless
sensor not directly connected to the electrically conductive
medium, measuring an AC voltage waveform shape; and, calculating
power usage at the destination node in response to the AC current
measurement, the measurement of the AC voltage waveform shape, and
an AC voltage potential.
2. The method of claim 1 wherein measuring the AC current includes
measuring the AC current at a first node located between the source
node and the destination node; wherein measuring the AC voltage
waveform shape includes measuring the AC voltage waveform shape at
the first node; and, wherein calculating the power usage includes
calculating the power usage in response to the AC voltage potential
selected from a group consisting of an AC voltage potential
estimate, an actual measurement of the AC voltage potential, and a
value supplied by an external source.
3. The method of claim 2 wherein measuring the AC current includes
measuring a plurality of instantaneous current values (I) each AC
cycle; wherein measuring the AC voltage waveform shape includes
measuring AC voltage phase; wherein calculating power usage
includes: calculating AC current phase and an AC current root mean
square (I.sub.RMS) value from the plurality of I values; finding a
difference angle (.theta.) between the phase of the AC current and
the phase of the AC voltage; and, using a V.sub.RMS value derived
from the AC voltage potential, finding cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.
4. The method of claim 2 wherein measuring the AC current includes
measuring a plurality of instantaneous current values (I) each AC
cycle; wherein measuring the AC voltage waveform shape includes
measuring AC voltage phase; wherein calculating power usage
includes: calculating AC current phase from the plurality of I
values; creating a synthesized AC voltage signal with a plurality
of instantaneous voltage values (V) each cycle, by scaling an AC
voltage magnitude by an estimate of the AC voltage waveform shape
having the same phase as the measured AC voltage waveform; finding
a difference angle (.theta.) between the phase of the AC current
and the phase of the AC voltage; using the difference angle to
maintain displacement between the I and V instantaneous values;
and, summing (I.times.V) over a plurality of values.
5. The method of claim 2 wherein measuring the AC current includes
measuring a plurality of instantaneous current values (I) each AC
cycle; wherein calculating power usage includes: calculating AC
current phase from the plurality of I values; creating a
synthesized AC voltage signal with a plurality of instantaneous
voltage values (V) each cycle, by scaling an AC voltage magnitude
by the measured AC voltage waveform shape; and, finding a
difference angle (.theta.) between the phase of the AC current and
the phase of the AC voltage; using the difference angle to maintain
displacement between the I and V instantaneous values; and, summing
(I.times.V) over a plurality of values.
6. The method of claim 2 wherein measuring the AC current includes
measuring a plurality of instantaneous current values (I) each AC
cycle; the method further comprising: high-pass filtering the
plurality of measured I values; wherein calculating power usage
includes: calculating a harmonic distortion power factor (hpf) from
the I values and filtered I values; finding a difference angle
(.theta.) between the phase of the AC current and the phase of the
AC voltage; and, using a V.sub.RMS value derived from the AC
voltage potential, finding cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.times.hpf.
7. The method of claim 2 further comprising: applying a known load
to a second node connected to the electrically conductive medium;
in response to measuring the AC current across the known load,
calculating V.sub.RMS; and, using V.sub.RMS as the AC voltage
potential in calculating power usage.
8. The method of claim 7 wherein applying the known load to the
second node includes applying the known load for sub-cycle
durations.
9. The method of claim 7 wherein applying the known load to the
second node includes applying the known load selected from a group
consisting of a resistor, a capacitor, and an inductor.
10. The method of claim 2 wherein measuring the AC current includes
measuring a plurality of instantaneous current values (I) each AC
cycle: wherein calculating power usage includes: calculating AC
current phase from the plurality of I values; using a plurality of
measured instantaneous voltage values (V) each cycle; finding a
difference angle (.theta.) between the phase of the AC current and
the phase of the AC voltage; using the difference angle to maintain
displacement between the I and V instantaneous values; and, summing
(I.times.V) over a plurality of values.
11. A method for calculating power using information measured at
different nodes, the method comprising: providing an electrically
conductive medium carrying alternating current (AC) electrical
current, associated with an AC voltage, from a source node to a
destination node; measuring the AC current through the electrically
conductive medium at a first node; measuring the AC voltage on the
electrically conductive medium at a second node; transmitting
measurement data from a test node selected from a group consisting
of the first node, the second node, and both the first and second
nodes, the transmitted measurement data including a measured
time-lag value, where the time-lag value is a duration of time
between when data is measured, and when the data is transmitted;
and, calculating power usage in response to the AC current
measurement, the AC voltage measurement, and the time-lag
value.
12. The method of claim 11 wherein transmitting data includes
transmitting data to a node selected from a group consisting of the
first node, the second node, and a collection node.
13. A system for calculating power using a contactless voltage
phase sensor, the system comprising: a contactless alternating
current (AC) current sensor configured to couple to a first node of
an electrically conductive medium carrying AC current, associated
with an AC voltage, from a source node to a destination node, the
contactless AC current sensor having an output for supplying an AC
current measurement; a contactless voltage phase sensor configured
to couple to the first node without directing connecting to the
electrically conductive medium, and having an output for supplying
an AC waveform shape measurement; and, a calculator having an input
port the accept the AC current measurement and the AC waveform
shape measurement, the calculator having an output to supply a
power usage value for the destination node in response to the AC
current measurement, the AC voltage waveform shape measurement, and
an AC voltage potential.
14. The system of claim 13 wherein the calculator calculates the
power usage in response to the AC voltage potential selected from a
group consisting of an AC voltage potential estimate, an actual
measurement of the AC voltage potential, and a value supplied by an
external source.
15. The system of claim 14 wherein the AC current sensor supplies a
plurality of instantaneous current values (I) each AC cycle;
wherein the voltage phase sensor supplies an AC voltage phase
measurement; and, wherein the calculator calculates AC current
phase and a AC current root mean square (I.sub.RMS) value from the
plurality of I values, finds a difference angle (.theta.) between
the phase of the AC current and the phase of the AC voltage, and
uses a V.sub.RMS value derived from the AC voltage potential to
find cos (.theta.).times.I.sub.RMS.times.V.sub.RMS.
16. The system of claim 14 wherein the AC current sensor supplies a
plurality of instantaneous current values (I) each AC cycle;
wherein the voltage phase sensor supplies an AC voltage phase
measurement; and, wherein the calculator calculates AC current
phase from the plurality of I values and creates a synthesized AC
voltage signal with a plurality of instantaneous voltage values (V)
each cycle, by scaling an AC voltage magnitude by an estimate of
the AC voltage waveform shape having the same phase as the AC
voltage phase measurement, the calculator finding a difference
angle (.theta.) between the phase of the AC current and the phase
of the AC voltage, using the difference angle to maintain
displacement between the I and V instantaneous values, and summing
(I.times.V) over a plurality of values.
17. The system of claim 14 wherein the AC current sensor supplies a
plurality of instantaneous current values (I) each AC cycle; and,
wherein the calculator calculates AC current phase from the
plurality of I values and creates a synthesized AC voltage signal
with a plurality of instantaneous voltage values (V) each cycle, by
scaling an AC voltage magnitude by the measured AC voltage waveform
shape, the calculator finding a difference angle (.theta.) between
the phase of the AC current and the phase of the AC voltage, using
the difference angle to maintain displacement between the I and V
instantaneous values, and summing (I.times.V) over a plurality of
values.
18. The system of claim 14 wherein the AC current sensor includes a
high-pass filter and supplies a plurality of instantaneous current
values (I) each AC cycle with a plurality of high-pass filtered I
values; wherein the voltage phase sensor supplies an AC voltage
phase measurement; and, wherein the calculator calculates AC
current phase from the plurality of I values and calculates a
harmonic distortion power factor (hpf) from the I values and
filtered I values, the calculator finding a difference angle
(.theta.) between the phase of the AC current and the phase of the
AC voltage, and using a V.sub.RMS value derived from the AC voltage
potential, to find cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.times.hpf.
19. The system of claim 14 further comprising: a known load
configured to selectively connect to a second node of the
electrically conductive medium; wherein the AC current sensor
supplies a calibrated AC current measurement when the known load is
connected; and, wherein the calculator calculates a V.sub.RMS value
in response to the known load and calibrated AC current
measurement, and uses the V.sub.RMS value as the AC voltage
potential in calculating power usage.
20. The system of claim 19 wherein the known load is selectively
connected to the second node for sub-cycle durations.
21. The system of claim 19 wherein the known load is selected from
a group consisting of a resistor, a capacitor, and an inductor.
22. The system of claim 14 wherein the AC current sensor supplies a
plurality of instantaneous current values (I) each AC cycle; and,
wherein the calculator accepts a plurality of measured
instantaneous voltage values (V) each cycle from the external
source, the calculator calculating AC current phase in response to
the plurality of I values, finding a difference angle (.theta.)
between the phase of the AC current and the phase of the AC
voltage, using the difference angle to maintain displacement
between the I and V instantaneous values, and summing (I.times.V)
over a plurality of values.
23. A system for calculating power using information measured at
different nodes, the system comprising: an alternating current (AC)
current sensor configured to couple to a first node of an
electrically conductive medium carrying AC current, associated with
an AC voltage, from a source node to a destination node, the AC
current sensor having an output for supplying AC current
measurement data; an AC voltage sensor configured to couple to a
second node, and having an output for supplying AC voltage
measurement data; wherein a test node selected from a group
consisting of the first node, the second node, and both the first
and second nodes, transmits measurement data including a measured
time-lag value, where the time-lag value is a duration of time
between when data is measured, and when the data is transmitted;
and, the system further comprising: a calculator having an input to
accept the AC current measurement data, the AC voltage measurement
data, and the time lag values, an output to supply destination node
power usage data in response to the inputs.
24. The system of claim 23 wherein the calculator is located at a
node selected from a group consisting of the first node, the second
test node, and a third node.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] This invention generally relates to alternating current (AC)
power measurement and, more particularly, to a system and method
for measuring AC power with a contactless voltage waveform shape
sensor.
[0003] 2. Description of the Related Art
[0004] The electric energy used in a residence is commonly measured
with electric meters that have full access to the electric power
conductors feeding that residence. However, the electric meter is
typically the property of the local electric utility and the
measurements made are generally not available to the residence by
automated means. In order for the resident to obtain electric
consumption data in near real-time, an additional electric meter is
often installed behind the utility's electric meter (at the
resident's side). The installation of an additional electric meter
generally requires access to the same electric conductors as the
utility owned meter. The connections consist of two types: 1) a
current sensor which is often made with a ferrite core transformer
placed around each conductor and 2) a physical electrical
connection to measure the voltage on each conductor.
[0005] Placement of the current sensors around the insulated
conductors requires no modification to wiring. However, tapping the
electrical conductors to make voltage measurements often requires
physically accessing dangerously high voltage conductors. For
safety reasons, the voltage measurement connections often require
the installation of permanent attachment points which increases the
cost of installation and the expertise required (often a licensed
electrician).
[0006] FIG. 14 is a drawing depicting a multi-family metering
system (prior art). In a multi-family installation, the number of
conventional meters and electrical connections required scales with
the number of residences metered. At each 240V split phase electric
meter, the following is required: [0007] a connection to neutral,
L1, and L2 [0008] current loops placed around L1 and L2
[0009] Thus, for 50 apartments, 150 connections and 100 current
loops would be required. However, since the meters are all likely
on one supply bus, L1, L2, and neutral connection (3 tap points)
are needed at the feed bus and then the voltages are delivered to
the meters (via an external voltage distribution wire network).
[0010] Conventionally, a voltmeter requires a contact connection,
and it's bundled with the current sensor and put in the panel; thus
requiring installation by an electrician. The power is computed
from the current and voltage measurements. The power measurements
are sent to a collector. The collector can show the measurements on
a server or it can send them to a display for the user to view.
Some new components described in the literature are non-contact
voltage meters using inexpensive electronics. One example is a
magnetometer plate for measuring current outside of the panel, and
a calibration load used for scaling of the amperage detected by the
magnetometer. A magnetometer is a device that measures the strength
or direction of a magnetic field. In this case, the flow of current
in the conductor induces a magnetic field which is picked up by the
magnetometer.
[0011] Non-contact single node monitors, such as the ones
advertised by Owl.RTM. monitor and the PowerSave EnviR, are easy to
install because they do not need to be wired in. The drawback is
that they need to be battery powered and they suffer inaccuracies
because of the poor voltage estimates. The power calculation they
use is: I.sub.rms.times.V.sub.rms. Current transformers provide
waveform, phase, and magnitude measurements, and a nominal
V.sub.rms is estimated. Distortions result as a result of using the
nominal V.sub.rms, and from the failure to account for harmonic
power factor, displacement power factor, and harmonic voltage.
[0012] FIG. 12 is a block diagram depicting a first power
measurement system (prior art). Recent papers describe non-contact
meters for single residences consisting of separate volt and
current nodes. One paper (Patel et al., "The Design and Evaluation
of an End-User-Deployable, Whole House, Contactless Power
Consumption Sensor", CHI 2010: Proceedings of ACM Conference on
Human Factors in Computing Systems, April 2010) describes a current
shape sensor or magnetometer that is attached to the outside of the
electric panel and calibrated. Placement of the sensor on the
electric panel is crucial and is guided by 2 LED lights on the
sensor. A calibrating unit cycles through a series of known loads
(10 W, 100 W, and 200 W) that are pulsed at 1 Hz. This system
suffers from the fact that it doesn't actively measure voltage, and
instead relies on the one calibration (which has an assumed power
factor of 1). This approach improves upon the commercial products
that use current sensors as it does not require physical access to
the conductors, but does not account for changes in power factor or
any changes in voltage. One difference between Patel's system and
existing commercial meters is the use of the calibrated
magnetometer instead of a current transformer, either of which can
be used for current sensing.
[0013] In summary, Patel discloses a single node system that
performs a power calculation of: I.sub.rms.times.V.sub.rms. The
sensor is a magnetometer to determine current that must be
calibrated, and a nominal V.sub.rms is estimated. Distortions in
power measurements occur as a result of using the nominal
V.sub.rms, and the failure to account for harmonic power factor,
displacement power factor, and harmonic voltage.
[0014] A single node system is also proposed by Nora (Maciej A.
Noras, "Solid state electric field sensor", Proc. ESA Annual
Meeting on Electrostatics 2011, and Maciej A. Noras. "Electric
field sensor based on a varactor diode/MIS/MOS structure" IEEE,
Proceeding from Industry Applications Society Annual Meeting,
2010). Noras' proposed single node monitor calculates power as: sum
(I.times.V). The sensors are current transformers and electrostatic
voltage meters. An electrostatic voltage meter determines the
voltage of a conductor by measuring the electrostatic field
generated by that voltage. The value of V.sub.rms is estimated.
While the distortions in measurement are limited to the voltage
estimate, the electrostatic voltage meter may consume too much
power to be packaged as a self-powered (e.g., battery operated)
node. This problem can only be ameliorated by measuring voltage
sporadically.
[0015] FIG. 13 is a block diagram depicting a second power
measurement system (prior art). Another paper (Schmid et al.,
"Meter Any Wire, Anywhere by Virtualizing the Voltage Channel",
BuildSys, Nov. 10, 2010) describes synchronizing voltage phase and
voltage values from the voltage-measuring node to the current node.
Schmid describes virtualizing the voltage by implementing separate
nodes (contact and non-contact). This allows a homeowner to install
it by putting the non-contact node in the panel, and then plugging
in the V.sub.node into any nearby wall socket. This synchronization
requirement is similar to the synchrophasors on high voltage lines
that send voltage and current samples with GPS synchronized
timestamps. This system requires that the current node do all the
computing, and it requires the node clocks to remain synchronized.
While this approach improves significantly over Patel, the
synchronization requires 2-way radios and fairly complex phase
locked loops, which impact reliability, complexity, and cost.
[0016] It would be advantageous if a more accurate means of
measuring power usage existed that used a contactless AC voltage
sensor.
SUMMARY OF THE INVENTION
[0017] Disclosed herein are a system and method that permits third
parties to inexpensively instrument single-family or multi-family
buildings (with grouped or physically distant utility meters) to
locally monitor electrical consumption. The system and method may
also be used to conveniently monitor sub-circuits within the same
residence or facility. The system installation requires no
connections at the panel or an electrician.
[0018] An important issue in power metering is the need to measure
current, voltage, and the phase difference between the two, thereby
allowing power factor (PF) to accurately be assessed for each unit.
The determination of PF aids in the accurate calculation of actual
power consumption.
[0019] Conventionally, accurate power measurement requires the
installation of individual third party meters on each metered
circuit, which is prohibitive in terms of both equipment and
installation costs. The third party meters require multiple
connections to the electrical wiring, significant physical space
for mounting, and a communications link to each device for
gathering of the acquired measurements. Other conventional
approaches calculate power by using non-contact current meters and
the assumption of a fixed voltage. These non-contact meters often
require synchronization and yield grossly inaccurate results.
[0020] Accordingly, a method is provided for calculating power
using a voltage waveform shape measurement from a contactless
sensor. An electrically conductive medium carries alternating
current (AC) electrical current, associated with an AC voltage,
from a source node to a destination node. The method measures the
AC current through the electrically conductive medium. Using a
contactless sensor not directly connected to the electrically
conductive medium, an AC voltage waveform shape is measured. The
power usage at the destination node is calculated in response to
the AC current measurement, the measurement of the AC voltage
waveform shape, and the AC voltage potential. For simplicity, the
AC current and AC voltage waveform shape may both be measured at a
first node located between the source node and the destination
node. The AC voltage potential used in the power usage calculation
may be an estimate, an actual measurement, or a value supplied by
an external source (e.g., the power utility).
[0021] In one aspect, a plurality of instantaneous current values
(I) are measured each AC cycle, and the AC voltage phase is
determined from the AC voltage waveform shape. Then, power usage is
determined by calculating AC current phase and an AC current root
mean square (I.sub.RMS) value from the plurality of I values. A
difference angle (.theta.) is found between the phase of the AC
current and the phase of the AC voltage. Using a V.sub.RMS value
derived from the AC voltage potential,
cos(.theta.).times.I.sub.RMS.times.V.sub.RMS is found. In a
variation, AC current phase is calculated from the plurality of I
values. A synthesized AC voltage signal is calculated with a
plurality of instantaneous voltage values (V) each cycle, by
scaling an AC voltage magnitude by an estimate of the AC voltage
waveform shape (e.g., a sine wave) having the same phase as the
measured AC voltage waveform. A difference angle (.theta.) is found
between the phase of the AC current and the phase of the AC
voltage. The difference angle is used to maintain displacement
between the I and V instantaneous values, and (I.times.V) is summed
over a plurality of values. As another alternative, the synthesized
AC voltage signal can be created by scaling an AC voltage magnitude
by the measured AC voltage waveform shape.
[0022] Additional details of the above-described method and a
system for calculating power using a contactless voltage phase
sensor are provided below.
BRIEF DESCRIPTION OF THE DRAWINGS
[0023] FIG. 1 is a schematic block diagram of a system for
calculating power using a contactless voltage phase sensor.
[0024] FIG. 2 is a schematic block diagram depicting a harmonic
distortion power factor variation of the system of FIG. 1.
[0025] FIG. 3 is a schematic block diagram depicting a load
calibration variation of the system of FIG. 1.
[0026] FIGS. 4A through 4C are schematic diagrams depicting a
system for calculating power using information measured at
different nodes.
[0027] FIGS. 5A and 5B are schematic block diagrams depicting a
multi-residence power measurement system.
[0028] FIG. 6 is a block diagram contrasting a single node system
with a two node system.
[0029] FIG. 7 is a block diagram depicting a power factor
compensating system.
[0030] FIG. 8 is a block diagram depicting a voltage waveform shape
correcting power measurement system.
[0031] FIG. 9 is a block diagram depicting a power measurement
system with harmonic power factor correction.
[0032] FIG. 10 is a flowchart illustrating a method for calculating
power using a voltage waveform shape measurement from a contactless
sensor.
[0033] FIG. 11 is a flowchart illustrating a variation of the
method for calculating power using information measured at
different nodes.
[0034] FIG. 12 is a block diagram depicting a first power
measurement system (prior art).
[0035] FIG. 13 is a block diagram depicting a second power
measurement system (prior art).
[0036] FIG. 14 is a drawing depicting a multi-family metering
system (prior art).
DETAILED DESCRIPTION
[0037] FIG. 1 is a schematic block diagram of a system for
calculating power using a contactless voltage phase sensor. The
system 100 comprises a contactless alternating current (AC) current
sensor 102 configured to couple to a first node 104 of an
electrically conductive medium 106 carrying AC current, associated
with an AC voltage. The first node 104 may be located at any point
between the source node 108 and the destination node 110. The
system 100 may be enabled using any conventional current sensor.
For example, a sensor made with a ferrite core transformer measures
inductance generated by current, and once calibrated, can yield a
current measurement. Alternatively, the AC current sensor may be of
the type that actually contacts the conductor. For example, a
current meter connected in series may be used, but then the full
current runs through the meter. It is more typical to put a shunt
resistor in series and then measure the voltage drop across the
shunt resister. This makes the measuring equipment more manageable.
Also the shunt resistor is sized for the expected loads and then a
standard voltmeter can be used to detect the current.
[0038] The electrically conductive medium 106 is connected from the
source node 108 to the destination node 110. Typically, the
electrically conductive medium is copper wires, with an AC voltage
potential between wires L1 and L2 (not shown), and a
neutral/ground. The contactless AC current sensor 102 has an output
on line 112 for supplying an AC current measurement. A contactless
voltage phase sensor 114 is configured to couple to the first node
104 without directly connecting to the electrically conductive
medium 106. The voltage phase sensor 114 has an output on line 116
for supplying an AC waveform shape measurement. In one aspect, the
voltage phase sensor is a flat plate, or a plate curved to at least
partially surround the electrical conductive medium. It is not
necessarily important where the single wire carrying the
measurement signal connects to the plate 114. The plate picks up a
charge from the conductor 106. That charge may then be applied to
an A/D converter to make the measurements.
[0039] A calculator 118 has an input port on line 112 to accept the
AC current measurement and an input port on line 116 to accept the
AC waveform shape measurement. The calculator 118 has an output on
line 120 to supply a power usage value for the destination node in
response to the AC current measurement, the AC voltage waveform
shape measurement, and an AC voltage potential. Here, the
calculator is shown as a separate node in the system. Alternatively
however, the calculator may be located at the first node, the
source node, the destination node, or a node making an AC voltage
potential measurement. The calculator 118 calculates the power
usage in response to an AC voltage potential estimate (e.g., stored
in memory 122), or an actual measurement of the AC voltage
potential. As shown, the source node in this example supplies the
measured AC voltage potential on line 124, however, the measurement
may be made at other nodes in the system, The AC voltage potential
may also be a value supplied by an external source on line 126
(e.g., from the power utility).
[0040] In one aspect, the AC current sensor 102 supplies a
plurality of instantaneous current values (I) each AC cycle. The
voltage phase sensor 114 supplies an AC voltage phase measurement.
For example, the voltage axis zero crossings of the voltage
waveform shape, and the interval between zero crossings, can be
used to yield the voltage phase. The calculator 118 calculates AC
current phase and an AC current root mean square (I.sub.RMS) value
from the plurality of I values, and finds a difference angle
(.theta.) between the phase of the AC current and the phase of the
AC voltage. The calculator uses a V.sub.RMS value derived from the
AC voltage potential to find cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS. As noted above, the AC
voltage potential may be an estimate, and actual measured value, or
a value supplied by an external source. The phase value .theta. may
be adjusted by a fix value set to compensate for a known phase
offset.
[0041] In a related variation, the calculator 118 calculates AC
current phase from the plurality of I values. The calculator 118
creates a synthesized AC voltage signal with a plurality of
instantaneous voltage values (V) each cycle, by scaling an AC
voltage magnitude (derived from the AC voltage potential) by an
estimate of the AC voltage waveform shape having the same phase as
the AC voltage phase measurement. In the simplest form the AC
waveform may be estimated as a perfect sine wave, however, other
waveform shape estimates are possible. The calculator 118 finds a
difference angle (.theta.) between the phase of the AC current and
the phase of the AC voltage, and uses the difference angle to
maintain displacement between the I and V instantaneous values.
Finally, the calculator sums (I.times.V) over a plurality of
values. Although not explicitly shown, the AC current phase may
alternatively be determined by the current sensor.
[0042] In another related variation, the calculator 118 calculates
AC current phase from the plurality of I values and creates a
synthesized AC voltage signal with a plurality of instantaneous
voltage values (V) each cycle, by scaling an AC voltage magnitude
by the measured AC voltage waveform shape. The calculator 118 finds
a difference angle (.theta.) between the phase of the AC current
and the phase of the AC voltage, uses the difference angle to
maintain displacement between the I and V instantaneous values, and
sums (I.times.V) over a plurality of values.
[0043] In another variation, the AC current sensor 102 supplies a
plurality of instantaneous current values (I) each AC cycle. The
calculator 118 accepts a plurality of measured instantaneous
voltage values (V) each cycle from the external source on line 126.
The calculator 118 calculates AC current phase in response to the
plurality of I values, and finds a difference angle (.theta.)
between the phase of the AC current and the phase of the AC
voltage. The calculator 118 uses the difference angle to maintain
displacement between the I and V instantaneous values, and sums
(I.times.V) over a plurality of values.
[0044] FIG. 2 is a schematic block diagram depicting a harmonic
distortion power factor variation of the system of FIG. 1. In this
aspect the AC current sensor 102 includes a high-pass filter 200
and supplies a plurality of instantaneous current values (I) each
AC cycle on line 112a with a plurality of high-pass filtered I
values on line 112b. The voltage phase sensor 114 supplies an AC
voltage phase measurement. The calculator 118 calculates AC current
phase from the plurality of I values and calculates a harmonic
distortion power factor (hpf) from the I values and filtered I
values. The calculator 118 finds a difference, angle (.theta.)
between the phase of the AC current and the phase of the AC
voltage, and uses a V.sub.RMS value derived from the AC voltage
potential, to find cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.times.hpf.
[0045] FIG. 3 is a schematic block diagram depicting a load
calibration variation of the system of FIG. 1. A known load 300 is
configured to selectively connect to (i.e. draw power from) a
second node (as shown, the destination node 110 is the second node)
of the electrically conductive medium 106. The AC current sensor
102 supplies a calibrated AC current measurement when the known
load is connected. The calculator 118 is able to calculate a
V.sub.RMS value in response to the known load and calibrated AC
current measurement, and it uses the V.sub.RMS value as the AC
voltage potential in calculating power usage. In one aspect, the
known load 300 is selectively connected to the second node 110 for
sub-cycle durations. The known load may be a resistor (or range of
resistors), a capacitor (or range of capacitors), an inductor (or
range of inductors), or a combination of the above-listed
elements.
[0046] FIGS. 4A through 4C are schematic diagrams depicting a
system for calculating power using information measured at
different nodes. The system 400 comprises an alternating current
(AC) current sensor 102 configured to couple to a first node 104 of
an electrically conductive medium 106 carrying AC current,
associated with an AC voltage. The electrically conductive medium
carries current from a source node 108 to a destination node 110.
The AC current sensor 102 has an output on line 112 for supplying
AC current measurement data.
[0047] An AC voltage sensor 402 is configured to couple to a second
node (shown here as the source node 108), and has an output on line
404 for supplying AC voltage measurement data. Note: the AC voltage
sensor may alternately be located at the first node or destination
node. Also note: in this aspect the AC current and AC voltage
sensors may be contactless or of a type that make contact with the
electrical conductor 106. A test node, which may be the first node
104, the second node 108, and both the first node 104 and second
node 108, transmits measurement data including a measured time-lag
value. The time-lag value is a duration of time between when data
is measured, and when the data is transmitted. A calculator 118 has
an input on line 112 to accept the AC current measurement data and
an input on line 404 the AC voltage measurement data, with the time
lag values. The calculator 118 has an output on line 120 to supply
destination node power usage data in response to the inputs.
[0048] FIG. 4A depicts the test node being the first node 104, with
the calculator being located at the second node. In other words,
the AC current sensor 102 transmits current data to the calculator
118, with a time-lag value. FIG. 4B depicts the test node being the
second node, with the calculator 118 being located at the first
node. In this aspect the AC voltage sensor 402 transmits AC voltage
data with a time-lag value. FIG. 4C depicts the AC current sensor
102 and AC voltage sensor 402 both acting as test nodes, with the
calculator being located at a third node 406. That is, both the AC
current sensor 102 and AC voltage sensor 402 send data with
time-lag values. The importance of the time-lag values in deriving
accurate power measurements is presented in more detail below.
[0049] FIGS. 5A and 5B are schematic block diagrams depicting a
multi-residence power measurement system. This system 500 measures
electrical power using a single voltage measurement node 502 to
obtain an accurate reading of the electric supply voltage levels.
Current flow measurements are obtained using separate current nodes
that obtain current sampled measurements via current transducers
(current transformers), or current magnitude and current phase
measurements. In one aspect not shown, the RMS voltage for a given
period is transmitted to the current node 504 via wireless
communications for computing of the power values. Alternatively as
shown, current flow and voltage phase measurements are transmitted
from the current node 504 to the voltage node 502. At the same
current node 504, voltage wave shape measurements are obtained for
N residences with non-contact voltage probes, such as electrostatic
voltage sensors. In one aspect, the voltage wave shape may be
scaled with the voltage level measurement from the voltage node,
and voltage samples are multiplied by current samples to obtain the
power. In other aspect, power is derived by computing the phase
difference between the voltage and the current waves
[0050] The remote current and voltage-shape measuring node 504 may
be powered by a long lasting battery or may harvest their power by
inductive coupling to the electric conductors of the measured
circuit. The voltage node 502 can tap into a breaker panel or into
any outlet, preferably close to the electric panel to avoid
distortions in voltage measurements during high current flows.
Computed values are available to be sent to other systems, such as
energy displays and energy management systems.
[0051] The number of physical connections for voltage sensing is
greatly reduced, thus saving equipment and installation costs. The
voltage sensing connections required for accurate voltage metering
are only a single set (L1, L2, and Neutral) per metering site. The
precision voltage measuring equipment is consolidated into a single
device which contains circuitry to make the voltage readings from
the single points on L1, L2, and N.
[0052] The devices used to make measurements of current and
voltage-shape measurements require no physical contact with the
electrical conductors. The sensors for current magnitude and phase
consist of current transformers which may be "split-core"
transformers which are opened and snapped closed around the
insulated electrical conductors. The voltage sensors at the
measured electric circuit locations do not require absolute
magnitude but only the voltage-shape. The voltage-shape
measurements are made with electrostatic sensors which, again, do
not require a physical connection with the electric conductors.
Rather, the voltage phase sensor need only be located near the
electric conductor being measured, such as placing the
electrostatic voltage sensor on the surface of the insulated
electrical conductor.
[0053] As the remote current and voltage-shape measuring devices
make no direct electrical connection with the circuit being
measured, power for the device may be supplied by a local battery.
In addition, the device may also harvest power from the electric
circuit being tested. When current flows through the electric
conductors (current transformer primary winding), the current
transformers may extract power from the conductor (at its secondary
winding) and store that power in local storage to then power the
device. Alternatively, the power may be harvested with separate
transformers for that purpose alone. Since there is no guarantee
that power will flow through the conductors being measured, the
power budget of the current/voltage-shape node may be managed by
means such as duty cycle adjustments to maintain communications
with the voltage node 502 to match the system requirements (e.g.
years of operation on an unused circuit). The current sensing,
voltage-shape sensing, and power harvesting pick-ups may be
constructed into a single attachment mechanism such that
installation around each electrical conductor is simplified.
[0054] In normal installations where the voltage magnitude and
phase angles of the supply conductors do not vary significantly,
only one voltage supply line may be contacted directly by the
voltage measuring device. The voltages on the other supply lines
may be estimated by the value read from the single sensed line. The
voltage phase of the other conductors may be estimated by adding a
fixed phase angle to the voltage phase of the sensed conductor. For
example, on a 240V, split phase supply with only connections to L1
and N, the voltage on L2 can be assumed to be approximately equal
to L1 and the voltage phase of L2 can be estimated to be the
voltage phase of L1 plus 180 degrees. This allows the installation
of the voltage sensing node to be plugged into a normal 120V AC
outlet, thus further simplifying installation (240V outlets being
less common).
[0055] Similar to the voltage node, the current/voltage-shape nodes
do not require the second voltage sensor and acquisition circuitry,
as the magnitudes of the voltages are approximately equal and the
phase offsets for the two phases are fixed. This reduction in
voltage sensing circuitry reduces the hardware costs and the power
budget for this device.
[0056] The current/voltage-shape node may compute the
root-mean-square power values (P.sub.1 and P.sub.2) for L1 and L2
by multiplying the A/D values captured from the current
transformers with the simultaneous A/D samples from the voltage
sensor that are by the RMS voltage received from the voltage node.
The phase of the voltage signal is not important at the voltage
node, but only the RMS value of the voltage A/D readings is
computed and transmitted from the voltage node to the
current/voltage-shape node.
[0057] At the current/voltage-shape node, the A/D converters on the
L1 and L2 current inputs provide the current samples, and the
electrostatic voltage sensor provides the voltage-shape samples.
These voltage-shape samples are multiplied by the scale factor (RMS
voltage from the voltage node/RMS value of the voltage-shape
samples). Power is computed by multiplying the current samples by
the scaled voltage-shape samples. Other values such as power factor
and energy consumption can be easily computed.
[0058] The measurements of RMS voltage at the single voltage node
can be applied to all readings at the current/voltage-shape nodes
as the voltage to all the metering locations is same. The voltage
loss across the service feeder is negligible due to the very low
line resistance of the feeder conductors. This calculated power
data may allow for energy usage visualization and personal
modification of energy usage. The data is also applicable for
automated energy management systems.
[0059] The main advantages of this invention over other electric
metering techniques are fewer electrical connections are required
to gain voltage samples reducing the cost of installation and
equipment. In practice, the voltage sample may be obtained from a
normal AC plug. The current/voltage-shape node has no electrical
connection for extremely easy installation. When power harvesting
is used at the current/voltage-shape node, the node has no
maintenance requirements (e.g. no battery change). The
current/voltage-shape nodes may use one way radio receivers to
receive data from the voltage node transmitters. This reduces power
consumption and cost at the current/voltage-shape node. Relatively
high precision is achieved in the power calculation as a result of
multiplying voltage and current samples. Lower precision is
achieved using RMS values and phase offset.
[0060] FIG. 6 is a block diagram contrasting a single node system
with a two node system. The bundling of all the sensors in a single
node can be distinguished from the distribution of sensors in two
nodes. A single node package may be assisted with a calibration
load node 600 that does not contain sensors and does not send data,
but rather adds known loads in patterns that can be recognized by
the sensors and can be used to calibrate the sensors.
[0061] The single node packages contain at least a current sensor
102, so it is called the current node 602 (or I.sub.node). The dual
node packages contain a second node that contains at least a
contact voltage sensor, so this node is called the voltage node 604
(or V.sub.node). Both systems use a collector node 606 (or
C.sub.node) to receive data and to send on the power values to a
display or a website. As shown, the power calculator 118 is located
in the I.sub.node 602, but it may alternatively be located in the
C.sub.node 606 or in the V.sub.node 604. Also, while the V.sub.node
604 is shown sending data to the I.sub.node 602, the system is not
limited to this particular signal path.
[0062] It is assumed that locations are wired with sufficiently
sized wire so that there is a negligible voltage drop between the
different circuits for any significant amount of time. If the
plug-in node (V.sub.node) is placed far away from the panel on a
circuit that draws large currents, there could be a voltage drop
that would distort the voltage measurements. For single residences,
a single voltage sensor is generally needed per phase. Most
residences use split-phase power and would therefore need two
voltage sensors. However, when only one voltage sensor is installed
on a first phase of a split-phase supply, the voltage of the second
phase may be estimated with reasonable accuracy by using the
voltage readings of the first phase.
[0063] A single current sensor is needed for each phase for every
circuit in the home to be monitored. For example, to monitor a
whole house two current sensors would be needed (one for each
phase). To separately monitor a 2-phase air conditioner, another
two current sensors would be needed. To separately monitor a
2-phase dryer another 2 current sensors would be needed. To
separately monitor the living room lights a single current sensor
would be needed, and so on. For calculating power in different
circuits, a shared voltage value may be used.
[0064] Multi-family residences typically share a single transformer
between clusters of residences, and all the residences on the same
transformer can share a single set of voltage meters. Given the
assumption above, a single pair (split-phase) of voltage sensors
can provide information needed for power calculations in all
residences attached to that transformer. For larger multi-family
residences, power may be distributed as 2 phases off of a 3-phase
transformer as in 208V WYE distribution. A single set of 3 voltage
sensors could then supply the voltage measurements for power
calculations in all residences attached to the various phases of
the transformer.
[0065] Voltage varies little in residential systems, both in
amplitude (specified at +-5V in a 120 v system in the US) and shape
(very close to a perfect sine wave). However, the phase of the
voltage relative to the phase of the current varies
substantially.
[0066] Measured and estimated voltage parameters can be
distinguished. Some conventional power monitors estimate the RMS
voltage to be a nominal value and depend only on a current meter to
calculate power. An improved power calculation can be made without
an actual contact voltage measurement. By combining an estimated
voltage value (perhaps entered by a user) with non-contact voltage
sensors that have been calibrated (as described below), the power
calculation can include the displacement power factor and thereby
provide a more accurate estimate of the actual power used.
[0067] A single cycle in a 60 Hz system takes 0.0166 seconds. In a
50 Hz system it takes 0.02 seconds. When making computations with
current and voltage waveforms, the raw values should not be shifted
in time significantly with respect to each other. For example, if
the voltage sensor is remote and the voltage value is sent via a
connection that adds a 1 ms delay, then the power factor
calculation could be off by 7%--this translates into a substantial
error in the power calculations. Similarly, a 2 ms delay would
induce a 27% error in the power factor calculation. The worst case
is where values travel over paths with variable delays, which could
introduce delays in a range. If one wireless remote current sensor
communicates via a router that introduces a 2 ms delay, the
resulting calculation could have an error of 27% or more.
Therefore, care must be taken to avoid accidental phase shifts
between readings. This may be accomplished by carefully calculating
the delays through the various paths, or by timestamping values
with accurate time values perhaps with micro-second accuracy.
[0068] In one aspect, the voltage waveform shape sensor can be
enabled using a simple copper bar taped onto a wire. The sensed
voltage shape and phase can be displayed, for example, on an
oscilloscope.
[0069] Real power can be calculated most accurately by summing the
current samples multiplied by the voltage samples over a period of
time (e.g. 4096 samples per cycle, 60 cycles per second). If
current and voltage samples come from separate meters, the samples
are sent to wherever the calculation is being performed. If one of
the components (voltage or current) is parameterized, an estimated
original waveform can be synthesized from the parameters in order
to access the samples.
[0070] If RMS values of current and voltage waveforms are used, the
"real power" is:
I.sub.rms.times.V.sub.rms.times.pf
[0071] Where pf is the power factor that includes contributions
from the phase shift between voltage and current and harmonics in
the current waveform (the voltage waveform is assumed to be
sinusoidal).
[0072] Note that the power company bills for "real power", but must
deliver complex power (VA) which is calculated as
I.sub.rms*V.sub.rms. These values can be very different. Suppose
the current is 10 Amps, the voltage is 120 Volts and the power
factor is 0.5. The "real power" is 600 Watts, however, the wiring
and equipment must be rated for 1,200 VA.
[0073] FIG. 7 is a block diagram depicting a power factor
compensating system. As noted above, one single node system is
disclosed herein and referred to as system 1.1 (see FIG. 1), where
the numeral "1" following the word "system" signifies a single node
system. System 1.1 calculates power as:
I.sub.rms.times.V.sub.rms.times.dpf, using an AC current sensor
that may be a current transformer or calibrated magnetometer to
measure the current waveform, magnitude, and shape, and a voltage
waveform shape sensor to measure V.sub.phase. A nominal V.sub.rms
may be estimated. Power measurement distortions may occur as a
result of using the nominal V.sub.rms, and a failure to account for
harmonic power factor and displacement power factor. Therefore, the
baseline calculation can be enhanced by adding a voltage phase
measurement to correct for displacement power factor. The
calculation consists of multiplying I.sub.rms.times.V.sub.rms, as
in the baseline calculation, but this result is then multiplied by
cos(angle difference between current and voltage). This system
corrects for displacement power factor, but does not correct for
harmonic power factor.
[0074] The power calculation can be performed in the I.sub.node or
in the C.sub.node (see FIG. 6). If power is calculated at the
C.sub.node, parameters are sent from the I.sub.node to the
C.sub.node and there is no data to synchronize.
[0075] As a variation to the above-described system, a voltage
phase estimator can be used to create a phase corrected voltage
waveform. Again, this system (system 1.2) may be enabled as a
single node system. Power is calculated as: sum (I.times.V). A
voltage waveform shape sensor is used to measure V.sub.phase, and a
current transformer or calibrated magnetometer provides waveform,
phase, and magnitude data for current. At the I.sub.node,
V.sub.phase and nominal V.sub.rms are estimated. Power measurement
distortions may result from using the nominal V.sub.rms, and from a
failure to account for voltage harmonics. However, compared to
system 1.1, distortion due to current harmonics is eliminated.
[0076] System 1.2 corrects for displacement power factor and
harmonic power factor due to current harmonics. The extra work
involved in this system consists of synthesizing the voltage wave
form and multiplying the current samples times the voltage samples.
This is similar to Schmid (see Background Section above), but the
voltage waveform is shifted according to the voltage phase
estimator. Unlike Schmid, V.sub.rms is not obtained from another
meter, but rather from scaling an ideal voltage waveform to a fixed
RMS value (e.g., 120V).
[0077] The power calculation can be performed in the I.sub.node or
in the C.sub.node. If power is calculated at the C.sub.node,
current samples and the phase difference between the voltage and
current are sent from the I.sub.node to the C.sub.node and there is
no data to synchronize.
[0078] FIG. 8 is a block diagram depicting a voltage waveform shape
correcting power measurement system. The system depicted in FIG. 8
(system 1.3) provides an assist with a voltage shape estimator to
create a shape corrected voltage waveform. Again, the system can be
enabled as a single node system, and power is calculated as: sum
(I.times.V). A voltage waveform shape sensor measures V.sub.shape,
and a current transformer or calibrated magnetometer provides
current waveform, phase, and magnitude data. In this system the
V.sub.shape is measured and a nominal V.sub.rms value is estimated.
Possible power measurement distortions may occur as a result of
using the nominal V.sub.rms, as voltage harmonics may occur in the
unsealed waveform.
[0079] A voltage waveform shape sensor is used instead of
synthesizing the voltage waveform. The voltage waveform shape
measurements simply need to be scaled appropriately, because while
the non-contact voltage waveform shape sensor gives a good estimate
of the waveform shape, the readings are attenuated by the
non-contact nature of the signal.
[0080] The calculation consists of multiplying the current samples
by the scaled voltage samples. This system corrects for voltage
waveform distortions. The extra work involved in this system
consists of scaling the voltage waveform and multiplying the
current samples times the voltage samples. This is similar to
Schmid, but the voltage waveform shape is measured, and a perfect
voltage sine wave is not assumed.
[0081] The power calculation can be performed in the I.sub.node or
in the C.sub.node. If power is calculated at the C.sub.node,
current samples and voltage shape samples are sent from the
I.sub.node to the C.sub.node and there is no data to
synchronize.
[0082] FIG. 9 is a block diagram depicting a power measurement
system with harmonic power factor correction. The system of FIG. 9
(system 1.4) assists in power calculations with the addition of a
fundamental frequency filter to add harmonic power factor. Again,
the system can be enabled as a single node system. Power is
calculated as: I.sub.rms.times.V.sub.rms.times.hpf. A voltage
waveform shape sensor measures V.sub.shape. A current transformer
or calibrated magnetometer provides current waveform, phase, and
magnitude data. In this system the V.sub.shape is measured and a
nominal V.sub.rms value is estimated. Possible power measurement
distortions may occur as a result of using the nominal V.sub.rms,
and from a failure to account for voltage harmonics and
displacement power factor.
[0083] A 60 Hz filter (or 50 Hz for international locales) is added
to the current sensor output, and the RMS value of this filtered
value is computed. This value is divided by the RMS value of the
unfiltered samples to obtain the harmonic power factor. The
calculation consists multiplying I.sub.rms.times.V.sub.rms as
before, and multiplying this result by the harmonic power factor.
This system corrects for harmonic power factor, but does not
correct for displacement power factor. The power calculation can be
performed in the I.sub.node or in the C.sub.node. If power is
calculated at the C.sub.node, parameters are sent from the
I.sub.node to the C.sub.node and there is no data to
synchronize.
[0084] The system shown in FIG. 9 can also be used to represent
system 1.5, a single node that calculates power as follows:
I.sub.rms.times.V.sub.rms.times.dpf.times.hpf. A voltage waveform
shape sensor measures V.sub.shape, and a current transformer or
calibrated magnetometer provides current waveform, phase, and
magnitude data. In this system the V.sub.shape is measured and a
nominal V.sub.rms value is estimated. Possible power measurement
distortions may occur as a result of using the nominal V.sub.rms,
and from a failure to account for voltage harmonics.
[0085] This system improves the power estimate by taking into
account both displacement and harmonic power factors. In theory,
this provides identical power values as multiplying current samples
by voltage samples, assuming the voltage is a sine wave. The power
calculation can be performed in the I.sub.node or in the
C.sub.node. If power is calculated at the C.sub.node, parameters
are sent from the I.sub.node to the C.sub.node and there is no data
to synchronize.
[0086] System 1.6 replaces the nominal voltage with a calibrated
voltage, and can be combined with systems 1.1, 1.2, 1.3, 1.4, or
1.5. System 1.6 can be enabled as a single node system with
multiple power calculation modes. A current transformer provides
current waveform, phase, and magnitude data. The voltage is
estimated by measuring V.sub.RMS at the I.sub.node. Distortions may
result from the above-mentioned voltage estimate.
[0087] In this system, the current sensor is a pre-calibrated
sensor so there is no need to calibrate the current sensor once
installed. The calibration node is then used to determine the
voltage RMS value. For example, the voltage is computed knowing the
resistor (R) values applied at the calibration node and the
corresponding current (I) associated with the R. V is computed by:
V=I.times.R.
[0088] System 1.7 replaces the nominal voltage with an external
voltage, it can be combined with systems 1.1, 1.2, 1.3, 1.4, or
1.5. System 1.7 can be enabled as a single node system with
multiple power calculation modes. A current transformer provides
current waveform, phase, and magnitude data. The voltage is
estimated by measuring V.sub.RMS at an external location and sent
to C.sub.node. Distortions may result from the above-mentioned
voltage estimate.
[0089] The current node meter sends its results to the C.sub.node.
Patel's calibration (see Background Section above) adds circuitry
at the I.sub.node that detects a calibration pattern introduced by
a calibrator plugged into an outlet on the circuit being measured.
The calibrator inserts and removes a resistor in a pattern. The
single node monitors the current sensor looking for that pattern to
determine if the changes in current are due to the calibration
routine. If so, then the power node performs extra calculations to
determine the voltage based on the current with and without the
known resistor.
[0090] Patel suggests a calibration pattern of inserting 10 W, 50
W, 100 W, 200 W resistors that are pulsed at 1 Hz. It is believed
that this setup is intended for occasional manual calibration, and
would be expensive because it needs large resistors to dissipate up
to 200 W. More practically, this pattern can be modified with
sub-cycle durations so that the measurements are taken at low
voltage intervals in the cycle. The calibration can be performed
automatically on regular intervals or continuously.
[0091] Alternatively, capacitors and/or inductors can be used
instead of resistors. This would introduce known values of reactive
power based on the inductance (L) and capacitance (C) introduced,
so real power would not be wasted (except for circuit
inefficiency). Lastly, if an electrostatic voltage shape meter is
to be calibrated, the above-mentioned techniques for the
calibrating a magnetometer can be applied.
[0092] In another aspect, a two node system may be used to measure
power. Such a system combines one non-contact node with a contact
V.sub.node. The non-contact node is easy to install because it
doesn't need to be wired in. The contact node can be wired in, or
can consist of a plug-in package (avoiding the need for an
electrician).
[0093] As noted above in the Background Section, Schmid describes a
possible setup where the V.sub.node parameterizes the voltage into
V.sub.rms and phase, and sends these wirelessly to the I.sub.node
together with timestamps for synchronization. The I.sub.node
creates a voltage waveform synthesized from the received info, and
then multiplies the samples together. This technique requires that
the nodes be time-synchronized.
[0094] The voltage waveform lends itself to being parameterized
because it is typically very close to a perfect sine wave (although
its phase, frequency, and amplitude vary). In contrast, the current
typically suffers from harmonics, so it is more difficult to
parameterize.
[0095] System 2.1 uses a voltage phase estimator to assist with
displacement power factor corrections. The numeral "2" following
the word "system" signifies a multi-node system This system is the
same as system 1.1 for the single node monitor, except that the
V.sub.rms is updated by communications means regularly from the
contact V.sub.node.
[0096] The power calculation can be performed in the I.sub.node,
the V.sub.node, or in the C.sub.node. If power is calculated at the
V.sub.node, parameters are sent from the I.sub.node to the
V.sub.node and there is no data to synchronize. If power is
calculated at the C.sub.node, parameters are sent from the sensor
nodes with no synchronization needed.
[0097] System 2.2 is the same as system 1.2 for the single node
monitor, except that the V.sub.rms is updated by a communications
means, regularly from the contact V.sub.node. System 2.2 uses a
voltage phase estimator to create a phase corrected voltage
waveform. The power calculation can be performed in the I.sub.node,
the V.sub.node, or in the C.sub.node. If power is calculated at the
V.sub.node, current samples and the phase difference from the
I.sub.node to the V.sub.node are sent, and the phase difference is
used to synchronize the samples. If power is calculated at the
C.sub.node, current samples, the phase difference from the
I.sub.node, to the C.sub.node, and the V.sub.node voltage samples
are sent to the C.sub.node. The phase difference is used to
synchronize the samples.
[0098] System 2.3 is the same as system 1.3 for the single node
monitor, except that the V.sub.rms is updated regularly by a
communications means from the contact V.sub.node. System 2.3 uses a
voltage shape estimator to create a shape corrected voltage
waveform. The power calculation can be performed in the I.sub.node,
the V.sub.node, or in the C.sub.node. If calculated at the
V.sub.node, current samples and the voltage shape samples from the
I.sub.node are sent to the V.sub.node and the phase difference are
used to synchronize the samples. If power is calculated at the
C.sub.node, current samples, the phase difference from the
I.sub.node to the C.sub.node, and V.sub.node voltage samples are
sent to the C.sub.node. The phase difference is used to synchronize
the samples.
[0099] System 2.4 is the same as system 1.4 for the single node
monitor, except that the V.sub.rms is updated regularly by a
communications means from the contact V.sub.node. System 2.4 uses a
fundamental frequency filter to correct for the harmonic power
factor.
[0100] System 2.5 combines systems 2.1 and 2.4, and is the same as
the system 1.5 for the single node monitor, except that the
V.sub.rms is updated regularly by a communications means from the
contact V.sub.node.
[0101] System 2.6 replaces time-synchronization with time-lags. If
the time delay introduced by sending the data between nodes is
known, then synchronized systems are not required. This system
eliminates the complexity of synchronizing clocks between the
nodes. Instead, the voltage phase arrives at the non-contact node
with a time-lag parameter. Even if the node clocks drift, the
difference in the time lag measurement is insignificant. This
system also does not require any new voltage sensors. The timing
reference between the sender and the receiver may be a transmitted
data packet (e.g., synchronized on the packet's preamble). The
sending node transmits the delay from its local data measurements
to the synchronizing transmission point (preamble) and the
receiving node measures the delay from the synchronizing reception
point (preamble) to the locally measured data.
[0102] System 2.7 is the same as system 2.2, except that the
instead of synthesizing a voltage wave form, samples are used from
the V.sub.node. System 2.7 uses a voltage phase estimator to create
a phase corrected voltage waveform.
[0103] The power calculation can be performed in the I.sub.node,
the V.sub.node, or in the C.sub.node. If power is calculated at the
I.sub.node, voltage samples are sent from the V.sub.node to the
I.sub.node and the samples are synchronized by adjusting the Vnode
samples by the phase difference. If power is calculated at the
V.sub.node, current samples and the phase difference from the
I.sub.node are sent to the V.sub.node and the phase difference is
used to synchronize the samples. If power is calculated at the
C.sub.node, current samples, the phase difference from the
I.sub.node to the C.sub.node, and the V.sub.node voltage samples
are sent to the C.sub.node. The phase difference is used to
synchronize the samples.
[0104] In most of these variations presented above, power can be
calculated at any node simply by sending information between nodes.
Thus, there is no reason to perform the calculations in the
I.sub.node. Especially in a multi-family site where there may be 50
I.sub.nodes (that may use batteries), the complexity and power
demands of the I.sub.nodes should optimally be reduced. Thus, the
current parameters can be sent to either the V.sub.node or to the
C.sub.node for processing
[0105] For dual node variations that send parameterized values, the
delay introduced by sending the parameters (instead of measuring
them locally) is insignificant. On the other hand, variations that
multiply samples must be carefully synchronized by either
synchronizing the node clocks and time-stamping the samples being
sent, or by tracking the time-lag introduced by the transmission.
Table 1 summarizes the system requiring synchronization.
TABLE-US-00001 TABLE 1 Power Synchronization Variation Calculation
Bandwidth needed Single node Sum(I .times. V) High bandwidth No
Single node Product of Low bandwidth No Parameters Dual node Sum(I
.times. V) High bandwidth Yes (or measure phase difference) Dual
node Product of Low bandwidth No Parameters
[0106] For variations using parameterized values. Wi-Fi or a simple
form of Power Line Communications (PLC) may be used. For variations
that multiply samples, either synchronization (clock or time-lags),
or measuring the phase difference can be used to shift the samples
appropriately.
[0107] Power for the non-contact nodes may be most easily supplied
with replaceable batteries with a long service life. However,
batteries do present a maintenance issue, especially when there are
many current measuring nodes, measuring a multitude of apartments
and/or sub-circuits. When current transformers are used to measure
the current flows, there is available power from those same
transformers. The current transformers may provide the dual
function of providing current measurements and power for the node.
Separate transformers are also possible for each function; power
and measurement.
[0108] Power can only be harvested from the measured circuit when
there is a current flow in that circuit. If the circuit has
inadequate power flow to keep the current measurement node supplied
with power, then the I.sub.node must adjust its duty cycle to match
the available power or be augmented with an additional power source
(e.g., batteries).
[0109] FIG. 10 is a flowchart illustrating a method for calculating
power using a voltage waveform shape measurement from a contactless
sensor. Although the method is depicted as a sequence of numbered
steps for clarity, the numbering does not necessarily dictate the
order of the steps. It should be understood that some of these
steps may be skipped, performed in parallel, or performed without
the requirement of maintaining a strict order of sequence.
Generally however, the method follows the numeric order of the
depicted steps. The method starts at Step 1000.
[0110] Step 1002 provides an electrically conductive medium
carrying AC electrical current, associated with an AC voltage, from
a source node to a destination node. Step 1004 measures the AC
current through the electrically conductive medium. Step 1006,
using a contactless sensor not directly connected to the
electrically conductive medium, measures an AC voltage waveform
shape. Step 1008 calculates power usage at the destination node in
response to the AC current measurement, the measurement of the AC
voltage waveform shape, and an AC voltage potential.
[0111] In one aspect, measuring the AC current in Step 1004
includes measuring the AC current at a first node located between
the source node and the destination node, and measuring the AC
voltage waveform shape in Step 1006 also includes measuring the AC
voltage waveform shape at the first node. Calculating the power
usage in Step 1008 includes calculating the power usage in response
to the AC voltage potential that is an AC voltage potential
estimate, an actual measurement of the AC voltage potential, or a
value supplied by an external source.
[0112] In one aspect, measuring the AC current in Step 1004
includes measuring a plurality of instantaneous current values (I)
each AC cycle. Measuring the AC voltage waveform shape in Step 1006
includes measuring AC voltage phase. Calculating power usage in
Step 1008 includes the following substeps. Step 1008a calculates AC
current phase and an AC current root mean square (I.sub.RMS) value
from the plurality of I values. Step 1008b finds a difference angle
(.theta.) between the phase of the AC current and the phase of the
AC voltage. Step 1008c, using a V.sub.RMS value derived from the AC
voltage potential, finds cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.
[0113] In a related variation. Step 1008d calculates AC current
phase from the plurality of I values. Step 1008e creates a
synthesized AC voltage signal with a plurality of instantaneous
voltage values (V) each cycle, by scaling an AC voltage magnitude
by an estimate of the AC voltage waveform shape having the same
phase as the measured AC voltage waveform. Step 1008b finds the
difference angle (.theta.) between the phase of the AC current and
the phase of the AC voltage. Step 1008f uses the difference angle
to maintain displacement between the I and V instantaneous values.
Step 1008g sums (I.times.V) over a plurality of values.
[0114] In another aspect, Step 1004 measures a plurality of
instantaneous current values (I) each AC cycle. Step 1008d
calculates AC current phase from the plurality of I values. Step
1008h creates a synthesized AC voltage signal with a plurality of
instantaneous voltage values (V) each cycle, by scaling an AC
voltage magnitude by the measured AC voltage waveform shape. Step
1008b finds the difference angle (.theta.) between the phase of the
AC current and the phase of the AC voltage. Step 1008f uses the
difference angle to maintain displacement between the I and V
instantaneous values. Step 1008g sums (I.times.V) over a plurality
of values.
[0115] In another aspect, Step 1004 measures a plurality of
instantaneous current values (I) each AC cycle. Step 1005 high-pass
filters the plurality of measured I values. Step 1008i calculates a
harmonic distortion power factor (hpf) from the I values and
filtered I values. Step 1008b finds a difference angle (.theta.)
between the phase of the AC current and the phase of the AC
voltage. Step 1008j, using a V.sub.RMS value derived from the AC
voltage potential, finds cos
(.theta.).times.I.sub.RMS.times.V.sub.RMS.times.hpf.
[0116] In another aspect, Step 1004 measures a plurality of
instantaneous current values (I) each AC cycle. Step 1008d
calculates AC current phase from the plurality of I values. Step
1008k uses a plurality of measured instantaneous voltage values (V)
each cycle. Step 1008b finds the difference angle (.theta.) between
the phase of the AC current and the phase of the AC voltage. Step
1008f uses the difference angle to maintain displacement between
the I and V instantaneous values. Step 1008g sums (I.times.V) over
a plurality of values.
[0117] In an another variation, Step 1007a applies a known load to
a second node connected to the electrically conductive medium. Step
1007b, in response to measuring the AC current across the known
load, calculates V.sub.RMS. Step 1007c uses V.sub.RMS as the AC
voltage potential in calculating power usage. In one aspect, Step
1007a applies the known load for sub-cycle durations. In another
aspect, Step 1007a applies the known load that may be a resistor,
capacitor, inductor, or a combination of the above-mentioned
elements.
[0118] FIG. 11 is a flowchart illustrating a variations in the
method for calculating power using information measured at
different nodes. The method begins at Step 1100. Step 1102 provides
an electrically conductive medium carrying AC electrical current,
associated with an AC voltage, from a source node to a destination
node. Step 1104 measures the AC current through the electrically
conductive medium at a first node. Step 1106 measures the AC
voltage on the electrically conductive medium at a second node.
Step 1108 transmits measurement data from a test node to the first
node, the second node, or both the first and second nodes. The
transmitted measurement data includes a measured time-lag value,
where the time-lag value is a duration of time between when data is
measured, and when the data is transmitted. Step 1110 calculates
power usage in response to the AC current measurement, the AC
voltage measurement, and the time-lag value. In one aspect, Step
1108 transmits data to the first node, the second node, or a
collection node.
[0119] A system and method have been provided for making power
measurements using a contactless voltage waveform shape sensor.
Examples of particular message structures and hardware units have
been presented to illustrate the invention. However, the invention
is not limited to merely these examples. Other variations and
embodiments of the invention will occur to those skilled in the
art.
* * * * *